thunlp / ERNIE

Source code and dataset for ACL 2019 paper "ERNIE: Enhanced Language Representation with Informative Entities"
MIT License
1.41k stars 267 forks source link

How to finetune bert base version on FIGER dataset? #88

Closed vincywong closed 2 years ago

vincywong commented 2 years ago

I want to test the FIGER dataset on bert base model,my command line is python3 code/run_typing_baseline.py \ --do_train --do_lower_case --data_dir data/FIGER --ernie_model bert-base-uncased \ --max_seq_length 256 --train_batch_size 32 --learning_rate 2e-5 --num_train_epochs 3.0 \ --output_dir output_figer_baseline --gradient_accumulation_steps 4 --threshold 0.3 --loss_scale 128 --warmup_proportion 0.2 But there is no "layer_types" item in bert-base model's configuration file. How to solve this problem?